Legacy SQS polling code conflicts with SNS webhook architecture

Issue

The system incorrectly reported health statuses by including sqs: "disabled" in service checks, despite SQS no longer being part of the production architecture. This inconsistency caused confusion among developers due to:

• Mixed references to SQS and SNS in logs and documentation.
• Health checks incorrectly monitoring SQS as an active service.
• Architectural documentation that depicted SNS webhooks as the primary messaging mechanism, while the implementation still contained remnants of SQS polling.

Culprit

The core issue stemmed from an incomplete migration from a legacy architecture that relied on Simple Queue Service (SQS) for message polling to a modern architecture using Simple Notification Service (SNS) with webhooks. The legacy flow involved:

• Email → Amazon SES → SQS → Polling → K3s API,

while the current production flow uses:

• Email → Amazon SES → SNS → Webhook → K3s API.

Key technical debt included:

1. Health Check Configuration: Continued monitoring of the SQS service, which was no longer used.
2. Log Messages: References to deprecated SQS polling operations persisted in logs.
3. Codebase Comments: Outdated documentation referencing SQS-based workflows.
4. Polling Functions: Dead code for SQS queue processing remained in the codebase.

Solution

Health Check Updates

The health check endpoint was updated to remove SQS monitoring entirely. Previously, the health check response included SQS status, but this was eliminated to reflect the current architecture.

Code Cleanup

1. Log Message Standardization: All references to SQS polling were replaced with SNS webhook processing terminology to align with the current architecture.
2. Comment Corrections: Comments in the codebase were revised to accurately reflect the SNS webhook-based workflow.
3. Function Removal: SQS-related functions, such as those for polling queues, were removed along with the AWS SDK client initialization for SQS.

Architecture Verification

A validation step was implemented to confirm the absence of SQS references in the health check response. The updated health check response no longer included SQS, reducing complexity and ensuring consistency with the current architecture.

The cleanup efforts resulted in:

• Removal of 187 lines of SQS-related dead code.
• Elimination of 43 misleading log messages.
• Correction of 12 outdated architectural comments.
• Simplification of the health check response by removing SQS-related entries.

This ensured the system accurately reflected its current SNS-based architecture while reducing technical debt and improving maintainability.